Rationale and design of the artificial intelligence scalable solution for acute myocardial infarction (ASSIST) study

Area of Science:

• Cardiology
• Artificial Intelligence in Medicine
• Medical Diagnostics

Background:

• Acute coronary syndrome (ACS) poses a significant health burden in Europe.
• Current diagnosis relies on clinical symptoms and ECG interpretation, which can be error-prone.
• ST-segment elevation on ECG is critical for reperfusion therapy but can be misleading.

Purpose of the Study:

• To improve the accuracy of ECG-assisted assessment for ACS patients.
• To reduce diagnostic and treatment delays in acute settings.
• To optimize and validate the Willem™ artificial intelligence platform for ECG analysis.

Main Methods:

• Retrospective observational study (ASSIST project) involving 10,309 patients.
• Utilizing the Willem™ AI platform to analyze 12-lead ECGs.
• Collecting ECG, clinical, and coronary angiography data for model optimization and validation.

Main Results:

• The study aims to evaluate the performance of the Willem™ platform in correctly identifying acute myocardial infarction with coronary artery occlusion.
• Internal model performance will be assessed using data from the retrospective study.
• External validation will be conducted in a subsequent stage.

Conclusions:

• The ASSIST project will generate data to refine the Willem™ platform for ECG-only myocardial infarction detection.
• The AI-driven approach is hypothesized to decrease diagnostic delays.
• Enhanced diagnostic accuracy and improved clinical outcomes are anticipated.